st: group-level reliability coefficient

[memsfeld@statssite.net]

Dear All,

My apologies for this being a very simple (I think!) question.  I am 
trying to determine the reliability of a scale for a group-level 
variable.  I have not used the loneway command before and want to make 
sure I know the correct steps to do this.

Here are the characteristics of data I'm using:

12 survey items (i.e., individual variables)
N = 200 "groups" (census tracts)   ("id_ngbd")
N = 1,000 respondents (They are unevenly grouped into the 200 census 
tracts).  ("id_pers")

I have aggregated individual responses to the 12 items at the level of 
the group (census tract) so that each variable now represents the mean 
score for each of the 200 neighborhoods.  Now, I want to create a 
summated scale of the 12 aggregated variables to represent the 
construct at the level of neighborhood.

To determine the reliability of this 12-item neighborhood-level scale, 
how to I specify the loneway command line? Do I list the 12 items 
making up the construct seperately?


Many thanks.
Margaret Emsfeld

________________________________________________________________________
Check out the new AOL.  Most comprehensive set of free safety and 
security tools, free access to millions of high-quality videos from 
across the web, free AOL Mail and more.
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/